[Rtk-users] Help Regarding FDK reconstruction

Tue Apr 19 09:14:24 EDT 2016

Sir,
I have changed origin for both projections and reconstructed volume. THERE
IS IMPROVEMENT... Sorry for my mistake, previously I have changed only for
projections.

my code modified as below
double origin_inp[Dimension] = { -25.775, -25.275, 0 }; // for projections

// for reconstructed volume.
origin[0] = -4.9422695280000006;
origin[1] = -4.1357903999999994;
origin[2] = -5.2317748560000004;

. Now I am facing another issue. My original dataset is 1500 projections.
But now I could work only with 150 projections. On exceeding this it shows
memory exception. "First-chance exception at 0x75DEC54F in finRecon.exe:
Microsoft C++ exception: itk::MemoryAllocationError at memory location
0x002CF6E0."
Occurs at vnl_ssc.h line 128.

My PC specification is:

Windows 7 professional
Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz
RAM 16.0GB
NVIDIA GeForce GT 730
Dedicated video memory: 4096 MB DDR3

Thanks and Regards,
AMJAD N

On Tue, Apr 19, 2016 at 5:22 PM, ghostcz <ghostcz at hotmail.com> wrote:

> int dim[Dime] = { 1032, 1012,1 };
>     double spacing_inp[Dime] = { 0.05, 0.05,0.05};
>     double origin_inp[Dime] = { 0, 0,0};
>
> This is wrong, but you said you tried with the offset right? i.e. moving
> the origin.
> Then check the orientation of your raw files maybe.
> I will try to run your code when I have some more time...
>
> L
>
>
>
> *From:* ghostcz <ghostcz at hotmail.com>
> *Sent:* Tuesday, April 19, 2016 1:39 PM
> *To:* AMJAD N <amjad.n.cet at gmail.com> ; rtk-users at public.kitware.com
> *Subject:* Re: [Rtk-users] Help Regarding FDK reconstruction
>
> Can you please upload the files again? The content has expired.
>
> L
>
> *From:* AMJAD N <amjad.n.cet at gmail.com>
> *Sent:* Tuesday, April 19, 2016 1:24 PM
> *To:* louie L <ghostcz at hotmail.com> ; rtk-users at public.kitware.com
> *Subject:* Help Regarding FDK reconstruction
>
> Sir,
>
> I haven't created geometry.xml. My code and data-set are attached in the
> previous discussion. Also parameters for generating geometry explained in
> the same thread. I am working without creating geometry.xml but creating
> geometry in the code itself.
>
> Thanks and Regards,
>
> AMJAD N
>
> On Tue, Apr 19, 2016 at 3:56 PM, louie L <ghostcz at hotmail.com> wrote:
>
>> Can you please send the geometry.xml?
>>
>> Louie
>>
>> Sent from my iOS
>>
>> Am 19.04.2016 um 12:15 schrieb AMJAD N <amjad.n.cet at gmail.com>:
>>
>> Sir,
>>
>> sdd and sid is measured from the experiment setup.
>> -0.5*spacing*(size-1) logic is applied and effect is same.
>>
>> The parameters we know are described below. Please go through it and help
>> me tackle this issue.
>>
>> Number of projections: 50
>>
>> Object size:
>> x = 9.924236 mm
>> y = 8.3048 mm
>> z = 10.505572mm
>>
>> Detector panel:
>> flat panel : Width: 1032 px., Height: 1012 px.
>> Width: 51.6mm Height: 50.6mm
>> (Detector pixel pith: 0.05mm)
>>
>> Source to Detector Distance = 459.154 mm
>> Source to object axis distance = 95.331mm
>>
>> Direction of rotation: clock-wise
>>
>> Angle
>> start: 0 degrees
>> increment: 7.2 degrees
>> total: 360 degrees
>>
>> Thanks and Regards.
>> AMJADN
>>
>> On Tue, Apr 19, 2016 at 11:32 AM, Simon Rit <
>> simon.rit at creatis.insa-lyon.fr> wrote:
>>
>>> Dear Amjad,
>>> I had a quick look at your dataset. For sure, it's a geometry problem.
>>> Do you have any information regarding the geometry? How did you obtain sdd
>>> / sid?
>>> Currently, I can see that you have set the projections and the volume
>>> origin to 0,0,0 which is not correct, you probably want to center them for
>>> a first test. If you want to center them, the formula to compute the origin
>>> is -0.5*spacing*(size-1) for each corrdinate.
>>> Simon
>>>
>>>
>>> On 18/04/2016 08:50, AMJAD N wrote:
>>>
>>> Hi, The data set is reattached below,. Someone please look into it and
>>> help me please,
>>>
>>> http://filenurse.com/download/bb2827a35cc99a14feee17d2c37cdf76.html
>>>
>>> Thanks in advance
>>> AMJAD
>>>
>>>>
>>>> Message: 1
>>>> Date: Thu, 14 Apr 2016 17:21:12 +0530
>>>> From: AMJAD N <amjad.n.cet at gmail.com>
>>>> To: rtk-users at public.kitware.com
>>>> Subject: [Rtk-users] Help Regarding FDK reconstruction
>>>> Message-ID:
>>>> <
>>>> mailto:CAHq1iunQyc-ZGRH7o6yrMtiwqnusAPnoKzRKPnxm5bpj_3b0%2BA at mail.gmail.com
>>>> <CAHq1iunQyc-ZGRH7o6yrMtiwqnusAPnoKzRKPnxm5bpj_3b0%2BA at mail.gmail.com>>
>>>> Content-Type: text/plain; charset="utf-8"
>>>>
>>>> Hi
>>>>
>>>> I am new to RTK. and I have been looking into it for the past two
>>>> weeks..I
>>>> think its an amazing tool..
>>>>
>>>> My problem is:
>>>>
>>>>
>>>> I have a set of 50 raw CT scan projections of an industrial fin as shown
>>>> (fin.jpg)
>>>> I am reading it using ImageSeriesReader with RawimageIO and been have
>>>> trying to reconstruct it using FDKConeBeamReconstructionFilter. Its been
>>>> almost a week that I am trying this,but
>>>> I could not reconstruct it correctly
>>>> I tried varying my geometry, origin, spacing etc ..but still no hope
>>>>
>>>> Can u please look into my program(finRecon.txt) and the current output
>>>> I am
>>>> getting is output_fin.mha , which is incorrect and suggest some help
>>>>
>>>> The link to my dataset:
>>>>
>>>> http://expirebox.com/download/b523394ea63c635d10171a69e687993b.html
>>>>
>>>> raw image dimensions : 1032 *1012
>>>> raw image 16 bit
>>>> current spacing: 0.05 0.05 0.05
>>>> origin: 0 0 0
>>>> step angle 7.2 degrees
>>>>
>>>>
>>>> Thanks in advance,
>>>>
>>>> AMJAD
>>>> -------------- next part --------------
>>>> An HTML attachment was scrubbed...
>>>> URL: <
>>>> http://public.kitware.com/pipermail/rtk-users/attachments/20160414/784c3452/attachment-0001.html
>>>> >
>>>>
>>>> ------------------------------
>>>>
>>>> Subject: Digest Footer
>>>>
>>>> _______________________________________________
>>>> Rtk-users mailing list
>>>> Rtk-users at public.kitware.com
>>>> http://public.kitware.com/mailman/listinfo/rtk-users
>>>>
>>>>
>>>> ------------------------------
>>>>
>>>> End of Rtk-users Digest, Vol 44, Issue 3
>>>> ****************************************
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Rtk-users mailing listRtk-users at public.kitware.comhttp://public.kitware.com/mailman/listinfo/rtk-users
>>>
>>>
>>>
>>
>> _______________________________________________
>> Rtk-users mailing list
>> Rtk-users at public.kitware.com
>> http://public.kitware.com/mailman/listinfo/rtk-users
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/rtk-users/attachments/20160419/90c0bd8b/attachment-0010.html>